The deal uses a special purpose vehicle, floor prices, and take-or-pay clauses to build a parallel rare earth supply chain that China cannot undercut — and it will force a market bifurcation within 18 months.
The U.S. Department of War just committed $750 million to a rare earth mine in Brazil that is already producing but whose output the Pentagon cannot touch until Q4 2026. The investment is a bet on a future the Pentagon is engineering in real time, a wager that sovereign-backed finance can build a supply chain China cannot undercut. The deal closed August 28, 2026, and it changes how the West buys the metals inside every precision weapon, fighter jet, and electric motor.
The $1.55 billion mechanism
The package totals $1.55 billion. The War Department’s Industrial Base Analysis and Sustainment program put in $750 million, $250 million more than originally planned. The Defense Logistics Agency committed to a forward purchase contract for at least $300 million over five years. A Tier-1 institutional bank added a $500 million senior secured revolving credit facility. The offtake agreement spans 15 years and covers all phase-one output of four magnet elements: neodymium, praseodymium, dysprosium, and terbium. It is structured as take-or-pay with floor prices.
The mine is Serra Verde’s Pela Ema deposit in Brazil. It entered commercial production in 2024 and will reach roughly 4,000 metric tons of rare earth oxide equivalent annually by the end of 2026. It is the only large-scale producer of all four magnetic rare earths outside Asia. A special purpose vehicle will buy 100 percent of that output. The U.S. government is the anchor customer. The bank provides liquidity. The War Department supplied the equity.
This is not a loan. It is sovereign-backed market creation.
Why the Pentagon became a banker
China controls roughly 90 percent of rare earth processing. The U.S. has tried to break that grip for decades. Mountain Pass shipped concentrate to China for separation. Lynas struggled to finance downstream capacity. The 2022 export scare, when Beijing threatened to cut off rare earth shipments, forced a reckoning. The Pentagon concluded that private capital alone could not build a parallel supply chain because Chinese producers can drop prices below any competitor’s cost of production and sustain the loss longer.
So the government stopped subsidizing projects and started buying output. The Industrial Base Analysis and Sustainment program became the primary tool. The U.S. committed $62.8 million to rare earth projects across four African countries days before the Brazil deal. The Trump administration is backing the Lobito Corridor railway in Africa, with the U.S. International Development Finance Corporation proposing up to $1 billion in financing. The Department of Energy proposed $10 million for seven critical minerals R&D projects on the same day. The pattern is unmistakable: the U.S. is using sovereign capital to build an entire logistics and processing corridor that private markets would not fund.
The SPV is a sovereign shield
The special purpose vehicle sits between the mine and the market, absorbing price risk. Serra Verde sells its entire Phase 1 output to the SPV at floor prices. The SPV then sells the material to the Defense Logistics Agency and, eventually, to commercial buyers. The take-or-pay clause forces the U.S. government to buy even if it does not need the material immediately, creating a strategic stockpile by default.
The floor prices are the critical detail. They guarantee Serra Verde’s profitability even if China dumps rare earths onto the spot market. The bank’s $500 million facility provides working capital so the SPV can pay the mine on time regardless of when the government draws down its purchase commitments. The War Department’s $750 million equity cushions the SPV against losses. This is a sovereign wealth fund structure applied to critical minerals. It shields the entire project from the market forces that killed every previous attempt to build non-Chinese rare earth capacity.
Thras Moraitis, CEO of Serra Verde Group, put it bluntly: “We are pleased that the SPV has achieved close under its DoW funding arrangements meaning the Offtake Agreement is now in force and in full effect enabling Serra Verde to leverage its unique capabilities and begin deliveries this year.” He added that “the US Government has substantially increased its funding of the SPV to $750 million and confirmed the $300 million in forward purchase contracts as well as signing a commitment letter for a further $500 million to be provided under a debt facility.”
The template for a parallel supply chain
The deal signals that the U.S. has accepted it cannot outcompete China on cost or speed in rare earth processing. The solution is sovereign-backed finance and multi-year offtake guarantees that create a parallel, geopolitically controlled supply chain. The first deliveries are expected early in the fourth quarter of 2026. That gives defense contractors, magnet makers, and EV supply chain managers roughly 18 months to reallocate sourcing from Chinese-dominated markets to a non-Asian alternative.
The U.S. will sign at least two more similar offtake agreements within 12 to 24 months. African and Australian rare earth projects are the most likely candidates. The $62.8 million already committed to African projects and the Lobito Corridor railway financing are the scaffolding for those deals. The Pentagon will expand the SPV structure to include downstream magnet production. Once the mine-to-magnet chain is fully sovereign-backed, the U.S. will have a complete, parallel rare earth supply chain by 2028.
China will respond. The most likely move is a price war. Beijing can lower rare earth export prices by 15 to 20 percent to undercut the floor prices in the U.S. offtake agreement. That would test the U.S. government’s willingness to absorb losses. The SPV structure is designed for exactly this scenario. The take-or-pay clause and floor prices mean the U.S. will keep buying at above-market rates while China dumps. The question is whether the political will holds when the cost becomes visible on a federal balance sheet.
If China does not cut prices within six months of the first SPV deliveries, the price war thesis is wrong. That would mean Beijing has decided that preserving market share is less important than maintaining margins, or that it cannot sustain a price war while its own domestic demand is growing. Either outcome is a strategic win for the U.S. The SPV structure wins either way: it either forces China to bleed cash or proves that a parallel supply chain can exist without a price war.
The consequence is a bifurcated market. Chinese rare earths will remain cheaper on the spot market. Western defense contractors and automakers will pay a premium for non-Chinese supply, locked in through long-term contracts. The spot price and the contract price will diverge. A two-tier market is the intended outcome. China’s dominance in rare earth processing becomes less strategically relevant because the U.S. is not trying to beat China on price. It is building a market where price is not the primary variable.
The $1.55 billion deal is a template, not an outlier. The structure can be replicated for lithium, cobalt, graphite, and other critical minerals. The SPV model turns the U.S. government into a de facto venture capitalist and anchor customer for any mine that produces materials the Pentagon deems strategic. The Defense Production Act provides the legal authority. The Industrial Base Analysis and Sustainment program provides the funding channel. The only remaining constraint is political appetite.
What you must do in the next 18 months
For defense contractors, magnet makers, and EV supply chain managers, the clock is running. The floor prices mean you will pay a premium for non-Chinese rare earths. Start auditing your supply chains now. Map every kilogram of neodymium and dysprosium back to its origin. The U.S. government is becoming both your customer and your competitor. The SPV structure means you cannot simply buy on the spot market; you need long-term contracts with the SPV or with mines that have similar offtake arrangements.
The risk is that China’s price war makes the floor price look expensive, and your CFO asks why you are paying above spot. The answer is that the U.S. government will absorb the loss to keep the supply chain alive. The opportunity is that early movers get preferential access to a new, geopolitically secure supply. Those who wait will find themselves at the back of a long queue.
The cost of independence
The $750 million bet on Serra Verde’s output is now the foundation of a new supply chain. The mine is already producing. The first SPV deliveries land in Q4 2026. The U.S. has decided the cost of independence is worth paying, even if it means losing a price war with China. The SPV structure insulates the mine from that war, but it does not insulate the U.S. taxpayer from the bill. The question is not whether the U.S. can build a non-China supply chain. It can. The question is whether it can sustain the losses long enough for the market to bifurcate permanently. The price war starts the day the first deliveries arrive.
